• Please note that there are a few updates and clarifications made in the Audiokarma Rules, mostly relating to advertising and the addition of the new "Paying it Forward" & "Giving back" forums in the AudioKarma Audio Marketplace section.

Hafler DH-200 smoking resister

Pro

Pro

New Member
Hi everybody, I'm new on the site and have read some very helpful articles. I picked up this Hafler amplifier and the left channel had 47.5 volts DC at the speaker terminals. The right side is fine with only a couple of millivolts DC. After replacing all the electrolytic caps and Transistors Q7, 8, 10 and 11, I still had the problem. Q9 which was spec'd as NP2222 appears unobtainable and I read that a bc547 was compatable, but after replacement and power up, R34 smoked.

Does anybody have experience with changing out NP2222? What did you use for this unobtainable transistor? What other problems do you anticipate with this left channel board?

Thank for your help.
 
Register to hide this ad
Pro said:
I read that a bc547 was compatable, but after replacement and power up, R34 smoked.
Click to expand...
Did you get the transistor installed..correctly? The BC547 is CBE. I'm not sure about the NP2222.
 
I think NP2222 is a manual mistake, it should be PN2222A which is the plastic (TO-92) equivalent of the TO-18 metal can 2N2222A.
Using your DMM in diode mode check all the transistors,diodes, with power off and big ecaps discharged of course :) you can check resistors too but sometimes they read different in value unless you lift a leg to isolate from the rest of the circuits.
You can check the fets for D-S shorts. gates should show and open to all other pins. It is railing so it could be Q14, &/or Q15
You can substitute a couple of 100 ohm 3W resistors for fuses F2,3R, as a means to test, so you do not smoke any more parts.
Another way to test is remove the fets and replace R31(220) with two 110 ohms in series, take the center point of the two 110ohms and connect to the feedback at terminal 11, this closes the FB loop and allows you to measure some DC voltages and test the amplifier. once you get your close to zero DC at terminal 11 then you can break the test connection, you can leave the two 110 in place if you ever need to test again.
Q3,4 act as constant current sources supplying around 1.2V/R12(560) = 2mA , split evenly between the collector resistors R6/7,17/19
Q9 is the Vbe multiplier and would develop anywhere up to around 2V across C-E for biasing the o/p fets.
Hope this helps you solve your problem.
 
Thank you both for the response.
Bluestime, the pn2222a is a much better replacement for the old piece and it is an EBC. Thank you.
Avionic, hopefully the smoked resistor is the extent of the damage! Thank you.
 
rcs16 said:
I think NP2222 is a manual mistake, it should be PN2222A which is the plastic (TO-92) equivalent of the TO-18 metal can 2N2222A.
Click to expand...
Thank you rcs16 for clarification on the PN2222A and the diagnostic plan. I will run it down.
 
r
rcs16 said:
You can check the fets for D-S shorts. gates should show and open to all other pins. It is railing so it could be Q14, &/or Q15
Click to expand...
Replaced the bad transistor with pn2222a and replaced the toasted diode.

With DMM negative on source and positive on drain, q 14,15,16,17 all show climb then slow drop even if I use the DMM to charge the gate. I'm guessing all are bad?
 
NP2222a.jpg
I cannot find any information on this transistor. I'm thinking with all the rebuilders of Hafler equipment, somebody has to know something! Your help is appreciated.
 
I still think it is a 2n2222a in a plastic package. Some used "NP" and some "PN" as a prefix I guess.
I have two DH-200's on my bench, I am modifying them with a new design DH-220C, they have PC-6A pcb's in them that I pulled out.
there are no NP2222A used. Q9 is a MPS8099. The manual that I have show them as NP2222
Q9 is the bias generator circuit, only 2-4V across C-E, so almost any npn will work in that location.
You have other problems. I could be the fets, have to troubleshoot.

Use this method,
first, using the ohmmeter, replace what you see or measure as damaged. sometimes you have to lift a resistor lead in order to isolate it from the circuit to measure properly.
You can check out one channel at a time, pull the 5A rail fuses that power up the other channel.
replace R31,(220/0.5W) with two (110/0.25W) resistors wired in series.
take the mid point between the two 110ohm resistors and make a temporary connection/wire to terminal #11, this completes the feedback loop for testing the pcb in isolation.
remove the mosfets. test using the DBT, if proper, no shorts, the lamp will be bright initially and dim down fast. if you have no fets installed the lamp will not even glow at power up because there is very little current being drawn.
measure all the voltages, right them down on a printed out schematic, post the results. compare the pcb voltages between channels, would have to install fuses on right channel for that.
once you verify that the pcb is op properly, you can remove the temp connection and put it back.
you can install only one set of mosfets for testing.

Good luck
 
Well some good news! I went through the unit and replaced faulty diodes, resistors, transistors, and 6 mosfets, probably from my ill-fated power up of the unit after improperly installing Q9 that started a lot of this ball rolling. And thanks rcs16, I installed a pn2222a in Q9. Cautiously with a DBT, and finally at full power, and it didn't blow up! Yikes! I can't believe I'm chalking this up to an accomplishment, but I'm taking it!

IMG_2648.jpg IMG_2649.jpg
There is obviously some kind of problem and imbalance here with right channel dc 0.380V and left channel 32.7V.. Thanks rcs16 for the diagnostic advice and the level of detail in your post. I feel like I am making progress, but still have a way to go here. Thanks for any advic eon this project.
 
Pro said:
There is obviously some kind of problem and imbalance here with right channel dc 0.380V and left channel 32.7V.. Thanks rcs16 for the diagnostic advice and the level of detail in your post. I feel like I am making progress, but still have a way to go here. Thanks for any advic eon this project.
Click to expand...

Some kind problem....you could say that!

Some kind of SERIOUS problem is more like it.

What ever you do, don't hook up any speakers to this amp.You should have about 10 millivolts DC per channel with a healthy dh200. You have 380 millitvolts and 32700 millivolts. The channel with 380 millivolts will probably not damage speakers, but it will sound like crap. The channel with 32700 millivolts will be lethal to speakers......and sound like crap.
 
Yeah guys I know...thanks. This amp was toast when I got it with other equipment in a package deal. I estimated that the repair cost of this piece vs just buying a modern high-current amp with some speaker protection on the used market would result in me placing this in the trash pile. It sits at the back of my bench and when I have a free moment from time to time, I try to figure it out, although quite honestly I'm losing patience with it. No worries. Thanks for all of the responses that have helped me get it at least to this point.
 
Sometimes the learning experience exceeds the value of the thing you fixed. I've definitely had things that had way too much effort and money into to justify otherwise.
 
gadget73 said:
Sometimes the learning experience exceeds the value of the thing you fixed. I've definitely had things that had way too much effort and money into to justify otherwise.
Click to expand...
That goes for the time spent, too. Fortunately many of us are learning even though it takes way too much effort, time, money to fix a unit to recover the costs. Fortunately, we can learn and take the lesson on to the next project. Like diving into a unit for the first time just figuring out how it comes apart can take some time. Next time not so much, but that was a simple example. I hope I'm learning a bit more than just that.
 
Do not throw it out, let me know if you are planning on that. I did give you instructions on how to test the driver board with the output stage mosfets disconnected, but I realize that it is a bit of fiddling, that you might not have the patients for. I have some old PC-6 boards that I do not need, since I have designed a new driver board. I could offer one to you but the parts cost to fix your boards would probably the same as my shipping costs to you. If the mosfets are toast they are expensive to buy new Exicon devices, some thing to way out of those costs vs getting a new amp.
 
You must log in or register to reply here.
Back
Top Bottom